Cyclosporin A, FK506, and rapamycin are microbial products with potent immunosuppressive properties that result primarily from a selective inhibition of T lymphocyte activation. Rapamycin was first described as an antifungal antibiotic extracted from a streptomycete (Streptomyces hygroscopicus) (Vezina et al. (1975) J. Antibiot., 28:721; Sehgal et al. (1975) J. Antibiot. 28:727; and Sehgal et al., U.S. Pat. No. 3,929,992). Subsequently, the macrolide drug rapamycin was shown to exhibit immunosuppressive as well as antineoplastic and antiproliferative properties (Morris (1992) Transplant Res 6:39-87).
Each of these compounds, cyclosporin A, FK506 and rapamycin, suppress the immune system by blocking distinctly different biochemical reactions which would ordinarily initiate the activation of immune cells. Briefly, cyclosporin A and FK506 act soon after Ca.sup.2+ -dependent T-cell activation to prevent the synthesis of cytokines important for the perpetuation and amplification of the immune response. Rapamycin acts later to block multiple affects of cytokines on immune cells including the inhibition of interleukin-2 (IL2)-triggered T-cell proliferation, but its antiproliferative effects are not restricted solely to T and B cells. Rapamycin also selectively inhibits the proliferation of growth factor-dependent and growth factor-independent nonimmune cells. Rapamycin is generally believed to inhibit cell proliferation by blocking specific signaling events necessary for the initiation of S phase in a number of cell types, including lymphocytes (Bierer et al. (1990) PNAS 87:9231-9235; and Dumont et al. (1990) J. Immunol 144:1418-1424), as well as non-immune cells, such as hepatocytes (Francavilla et al. (1992) Hepatology 15:871-877; and Price et al. (1992) Science 257:973-977). Several lines of evidence suggest that the association of rapamycin with different members of a family of intracellular FK506/rapamycin binding proteins (FKBPs) is necessary for the inhibition of G.sub.1 progression as mediated by rapamycin. For instance, the actions of rapamycin are reversed by an excess of the structurally FKBP-ligands FK506 or 506BD (Bierer et al. supra.; Dumont et al. supra.; and Bierer et al. (1990) Science 250:556-559).
Cyclosporin A binds to a class of proteins called cyclophilins (Walsh et al. (1992) J. Biol. Chem. 267:13115-13118), whereas the primary targets for both FK506 and rapamycin, as indicated above, are the FKBPs (Harding et al. (1989) Nature 341:758-7601; Siekienka et al. (1989) Nature 341:755-757; and Soltoff et al. (1992 J. Biol. Chem. 267:17472-17477). Both the cyclophilin/cyclosporin and FKBPI2/FK506 complexes bind to a specific protein phosphatase (calcineurin) which is hypothesized to control the activity of IL-2 gene specific transcriptional activators (reviewed in Schreiber (1991) Cell 70:365-368). In contrast, the downstream cellular targets for the rapamycin-sensitive signaling pathway have not been especially well characterized, particularly with regard to the identity of the direct target of the FKBP-rapamycin complex.
The TOR1 and TOR2 genes of S. cerevisiae were originally identified by mutations that rendered cells resistant to rapamycin (Heitman et al. (1991) Science 253:905-909) and there was early speculation that the FKBP/rapamycin complex might inhibit the cellular function of the TOR gene product by binding directly to a phosphoserine residue of either TOR1 or TOR2. Subsequently, however, new models for rapamycin drug interaction have been proposed which do not involve direct binding of the FKBP/rapamycin complex to the TOR proteins. For example, based on experimental data regarding cyclin-cdk activity in rapamycin treated cells, Stuart Schreiber wrote in Albers et al. (1993) J. Biol. Chem. 268:22825-22829:
"Although it is possible the TOR2 gene product is a direct target of the FKBP-rapamycin complex, a more likely explanation is that the TOR2 gene product lies downstream of the direct target of rapamycin and that the TOR2 mutation caused the protein to be constitutively active. If the latter model is correct, then the TOR2 gene product joins p70.sup.s6k, cyclin-dependent kinases, and cyclin D1 as proteins that lie downstream of the direct target of the FKBP-rapamycin complex and have been shown to play important roles in cell cycle progression. The identification of the direct target of the FKBP-rapamycin complex will likely reveal an upstream component of the signal transduction pathway that leads to G1 progression and will help delineate the signal transduction pathways that link growth factor-mediated signaling events and cyclin-cdk activity required for cell cycle progression."
Likewise, after studying the role of TOR1 and TOR2 mutations in rapamycin-resistant yeast cells, George Livi wrote in Cafferkey et al. (1993) Mol. Cell Biol. 13:6012-6023:
"Thus, the amino acid changes that we have identified in the rapamycin-released DRR1 [TOR1] protein may allow it to compensate for the loss of the proliferative signal inhibited by rapamycin by constitutively activating an alternative signal rather than by preventing its association with the FKBP12-rapamycin complex. The positions of the mutations within the kinase domain, but in a region not shared by the PI 3-kinases, support this idea. Therefore, it is entirely possible that DRR1 is not a component of the rapamycin-sensitive pathway in wild-type yeast cells. Instead, missense mutations in DRR1 at Ser-1972 may alter its normal activity and allow it to substitute for the function of an essential protein which is the true target of rapamycin."
It is an object of the present invention to identify cellular proteins which are the direct downstream target proteins for the FKBP/rapamycin complex, and isolate the genes encoding those proteins.